On 02-12-21, 05:43, Hongxing Zhu wrote:

quoted

-----Original Message-----
From: Lorenzo Pieralisi <redacted>
Sent: Wednesday, December 1, 2021 8:44 PM
To: Hongxing Zhu <hongxing.zhu@nxp.com>
Cc: l.stach@pengutronix.de; bhelgaas@google.com; Marcel Ziswiler
[off-list ref]; tharvey@gateworks.com;
kishon@ti.com; vkoul@kernel.org; robh@kernel.org;
galak@kernel.crashing.org; shawnguo@kernel.org;
linux-phy@lists.infradead.org; devicetree@vger.kernel.org;
linux-pci@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
linux-kernel@vger.kernel.org; kernel@pengutronix.de; dl-linux-imx
[off-list ref]
Subject: Re: [PATCH v6 0/8] Add the imx8m pcie phy driver and imx8mm
pcie support

On Thu, Nov 18, 2021 at 09:54:41AM +0800, Richard Zhu wrote:

quoted

Refer to the discussion [1] when try to enable i.MX8MM PCIe support,
one standalone PCIe PHY driver should be seperated from i.MX PCIe
driver when enable i.MX8MM PCIe support.

This patch-set adds the standalone PCIe PHY driver suport[1-5], and
i.MX8MM PCIe support[6-8] to have whole view to review this

patch-set.

quoted

The PCIe works on i.MX8MM EVK board based the the blkctrl power

driver

quoted

[2] and this patch-set. And tested by Tim and Marcel on the different
reference clock modes boards.

[1]

https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpa
tc

quoted

hwork.ozlabs.org%2Fproject%2Flinux-pci%2Fpatch%2F20210510141509.
929120

quoted

-3-l.stach%40pengutronix.de%2F&data=04%7C01%7Chongxing.zhu
%40nxp.c

quoted

om%7C3edb11e040e6412cf91108d9b4c85052%7C686ea1d3bc2b4c6fa9
2cd99c5c3016

quoted

35%7C0%7C1%7C637739594698843569%7CUnknown%7CTWFpbGZsb3d
8eyJWIjoiMC4wLj

quoted

AwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&
amp;sdata=

quoted

5xlZSnBYs1SIIbMnmlQwi0qtfLDgKbueLNjPWIPD1pw%3D&reserved=
0

quoted

[2]

https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpa
tc

quoted

hwork.kernel.org%2Fproject%2Flinux-arm-kernel%2Fcover%2F20210910
202640

quoted

.980366-1-l.stach%40pengutronix.de%2F&data=04%7C01%7Chon

gxing.zhu%

quoted

40nxp.com%7C3edb11e040e6412cf91108d9b4c85052%7C686ea1d3bc2
b4c6fa92cd99

quoted

c5c301635%7C0%7C1%7C637739594698843569%7CUnknown%7CTWFp
bGZsb3d8eyJWIjo

quoted

iMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%
7C3000&amp

quoted

;sdata=1388J8dLuKUc6KEUnWj5pLpkaPDC4kTIZFF%2BPTspHZY%3D&a

mp;reserved=0

quoted

Main changes v5 --> v6:
- Add "Reviewed-by: Rob Herring [off-list ref]" into #1 and #3

patches.

quoted

- Merge Rob's review comments to the #2 patch.

Main changes v4 --> v5:
- Set the AUX_EN always 1b'1, thus it can fix the regression introduced

in v4

quoted

  series on Marcel's board.
- Use the lower-case letter in the devicetreee refer to Marcel's

comments.

quoted

_ Since the default value of the deemphasis parameters are zero, only

set

quoted

  the deemphasis registers when the input paramters are none zero.

Main changes v3 --> v4:
- Update the yaml to fix syntax error, add maxitems and drop
description of phy
- Correct the clock name in PHY DT node.
- Squash the EVK board relalted dts changes into one patch, and drop

the

quoted

  useless dummy clock and gpio suffix in DT nodes.
- Add board specific de-emphasis parameters as DT properties. Thus

each board

quoted

  can specify its actual de-emphasis values.
- Update the commit log of PHY driver.
- Remove the useless codes from PCIe driver, since they are moved to
PHY driver
- After the discussion and verification of the CLKREQ# configurations

with Tim,

quoted

  agree to add an optional boolean property "fsl,clkreq-unsupported",

indicates

quoted

  the CLKREQ# signal is hooked or not in HW designs.
- Add "Tested-by: Marcel Ziswiler [off-list ref]" tag,

since

quoted

  Marcel help to test the v3 patch-set.

Main changes v2 --> v3:
- Regarding Lucas' comments.
 - to have a whole view to review the patches, send out the i.MX8MM

PCIe support too.

quoted

 - move the PHY related bits manipulations of the GPR/SRC to

standalone PHY driver.

quoted

 - split the dts changes to SOC and board DT, and use the enum instead

of raw value.

quoted

 - update the license of the dt-binding header file.

Changes v1 --> v2:
- Update the license of the dt-binding header file to make the license
  compatible with dts files.
- Fix the dt_binding_check errors.

Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.yaml    |   6

+++

quoted

Documentation/devicetree/bindings/phy/fsl,imx8-pcie-phy.yaml |  92

+++++++++++++++++++++++++++++++

quoted

arch/arm64/boot/dts/freescale/imx8mm-evk.dtsi                |

55 +++++++++++++++++++

quoted

arch/arm64/boot/dts/freescale/imx8mm.dtsi                    |

46 +++++++++++++++-

quoted

drivers/pci/controller/dwc/pci-imx6.c                        |

73 ++++++++++++++++++++++---

quoted

drivers/phy/freescale/Kconfig                                |

9 ++++

quoted

drivers/phy/freescale/Makefile                               |

1 +

quoted

drivers/phy/freescale/phy-fsl-imx8m-pcie.c                   |

237
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++++++++++++++++++++

quoted

include/dt-bindings/phy/phy-imx8-pcie.h                      |

14 +++++

quoted

9 files changed, 525 insertions(+), 8 deletions(-)

Hi Richard,

I can pull this series into the PCI tree (but not the dts changes that should
be routed elsewhere) or give an ACK for patch 8, please let me know
what's the best option.

[Richard Zhu] Hi Lorenzo:
First of all, thanks a lot for your kindly help.
To my original understand, #1-3 patch had been reviewed by Rob, might be
 merged into Rob's dt-binding git repos.
Shawn takes dts changes ( #4, #6 and #7). And PHY driver merged by vkoul
 or Kishon.
In the end, the PCIe changes(#8) are merged into PCIe git repos.

I'm appreciated if you pull this whole series although I'm not sure that
 you can do it or not.
Today, I'm glad to receive vkoul's comments about the PHY driver part.
Let me continue refine the PHY driver, send anther version of this patch-set.
Then, let's figure out what's the best option to merge this series.

I think phy binding and driver changes should go thru phy tree and pcie
binding and driver changes thru pcie tree, dt should be picked by
respective arch tree

-- 
~Vinod

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy